Skip to main content

Research Repository

Advanced Search

To Learn Image Super-Resolution, Use a GAN to Learn How to Do Image Degradation First

Bulat, Adrian; Yang, Jing; Tzimiropoulos, Georgios

Authors

Adrian Bulat

Profile Image

JING YANG JING.YANG@NOTTINGHAM.AC.UK
Assistant Professor

Georgios Tzimiropoulos



Abstract

This paper is on image and face super-resolution. The vast majority of prior work for this problem focus on how to increase the resolution of low-resolution images which are artificially generated by simple bilinear down-sampling (or in a few cases by blurring followed by
down-sampling). We show that such methods fail to produce good results when applied to real-world low-resolution, low quality images. To circumvent this problem, we propose a two-stage process which firstly trains a High-to-Low Generative Adversarial Network (GAN) to learn how to degrade and downsample high-resolution images requiring, during training, only unpaired high and low-resolution images. Once this is achieved, the output of this network is used to train a Low-to-High GAN for image super-resolution using this time paired low- and high-resolution images. Our main result is that this network can be now used to effectively increase the quality of real-world low-resolution images. We have applied the proposed pipeline for the problem of face super-resolution where we report large improvement over baselines and prior work although the proposed method is potentially applicable to other object categories.

Start Date Sep 8, 2018
Publication Date Oct 6, 2018
Publisher Springer Publishing Company
Pages 187-202
Book Title Computer Vision – ECCV 2018: 15th European Conference Munich, Germany, September 8–14, 2018 Proceedings, Part VI
ISBN 9783030012304; 9783030012311
APA6 Citation Bulat, A., Yang, J., & Tzimiropoulos, G. (2018). To Learn Image Super-Resolution, Use a GAN to Learn How to Do Image Degradation First. In Computer Vision – ECCV 2018: 15th European Conference Munich, Germany, September 8–14, 2018 Proceedings, Part VI, 187-202. https://doi.org/10.1007/978-3-030-01231-1_12
DOI https://doi.org/10.1007/978-3-030-01231-1_12
Keywords Image and face super-resolution; Generative Adversarial Networks; GANs
Publisher URL https://link.springer.com/chapter/10.1007/978-3-030-01231-1_12
Additional Information Conference Acronym: ECCV; Conference Name: European Conference on Computer Vision; Conference City: Munich; Conference Country: Germany; Conference Year: 2018; Conference Start Date: 8 September 2018; Conference End Date: 14 September 2018; Conference Number: 15; Conference ID: eccv2018; Conference URL: https://eccv2018.org/

Files





You might also like



Downloadable Citations

;